The Technology Vent and Rant Thread

[jbjb]

Originally Posted by ratinox

No, I get it. If salts are random/unique for each hash then there's no practical difference between direct brute force and generating a unique rainbow table for that hash, except that brute force should be faster. Rainbow tables are useful against unsalted hashes or when salts are reused. Then you get two or more for the price of one.

Exactly

[ownedbycats]

AI Overview tells me that the treatment of a certain medical condition varies based on its cause and severity.

I don't know what I would've done without this illuminating answer.

[Quoth]

Originally Posted by ownedbycats

AI Overview tells me that the treatment of a certain medical condition varies based on its cause and severity.

I don't know what I would've done without this illuminating answer.

Scraped from a medical website.

[ratinox]

Originally Posted by ownedbycats

AI Overview tells me that the treatment of a certain medical condition varies based on its cause and severity.

I don't know what I would've done without this illuminating answer.

I've had a few pet names for these things: regurgitive AI (because they regurgitate innumerable variations of the ingested corpus), derivative AI (because what they spew are derivative works, particularly in light of what Meta/Farcebook got caught at and how OpenAI are trying to get the courts to rule that their blatant, commercial piracy is fair use), but most recently "degenerative AI". Because as they feed on each other's droppings on das intarwebs, the results are akin to the degeneration and degradation seen in repeated VHS copying.

[Quoth]

Originally Posted by ratinox

…
Because as they feed on each other's droppings on das intarwebs, the results are akin to the degeneration and degradation seen in repeated VHS copying.

Brilliant.

[ownedbycats]

Originally Posted by ratinox

Because as they feed on each other's droppings on das intarwebs, the results are akin to the degeneration and degradation seen in repeated VHS copying.

19th generation - it is getting more difficult to follow what is happening on screen

Yeah no shit

[ownedbycats]

Seeing multiple articles referring to Warner Bros DVDs suffering from "laser rot." (1, 2, 3)

The term 'laser rot' refers to specifically one format, LaserDiscs - the issue is when the disc layers degrade, nothing to do with lasers. The more general term is disc rot.

Probably pedantic but it bothers me.

[JSWolf]

Originally Posted by ownedbycats

Yeah no shit

What I'd like to know is what happens if you copy the original VHS to S-VHS and then make S-VHS to S-VHS copies.

[ownedbycats]

Originally Posted by JSWolf

What I'd like to know is what happens if you copy the original VHS to S-VHS and then make S-VHS to S-VHS copies.

Here you go:

[Comfy.n]

What about D-VHS?

[ratinox]

Originally Posted by Comfy.n

What about D-VHS?

If it's digital to digital, like duping a CD to MD (MiniDisc) via TOSLink, with no analog conversion, then the copy should be 1:1 identical (less any SCMS stripping that might occur).

If copying requires decoding to analog TV signal and then digitizing this signal for recording then there will be some degradation depending on the quality of the digitizer and of the analog interconnects. Additional degradation will occur if lossy encodings are used -- same effect you get from re-encoding a JPEG as a JPEG.

[JSWolf]

Originally Posted by ratinox

If it's digital to digital, like duping a CD to MD (MiniDisc) via TOSLink, with no analog conversion, then the copy should be 1:1 identical (less any SCMS stripping that might occur).

If copying requires decoding to analog TV signal and then digitizing this signal for recording then there will be some degradation depending on the quality of the digitizer and of the analog interconnects. Additional degradation will occur if lossy encodings are used -- same effect you get from re-encoding a JPEG as a JPEG.

CD > MiniDisc > CD will result in a loss of data. Any stripping is not 1:1. You can even have a difference going from CD > CD unless you have the correct offset for the drive you are using.

[Quoth]

CD is uncompressed. Minidisc uses ATRAC compression. So not 1:1

CD to CD is only 1:1 for CD-ROM, not CD-Audio as audio has no error correction, though mostly the results are better than analogue tape and very many more generations are possible.

D-VHS uses compression, so any normal copying will multiply compression artefacts. Won't be as a bad as analogue VHS.

I don't think the video example is real because VHS has noise, drop-out and head switching, so degrades faster than that. NTSC is worse than PAL. Even original VHS recording has noticeable colour smearing on many machines on 1st recording as the colour bandwidth is much poorer than PAL and the time delay may not be accurate. The 3.57 MHz (NTSC) or 4.43 MHZ (PAL, SECAM) is converted to about 768 kHZ carrier recorded separately to the mono Luminance FM carrier. Called "Color under" and was originally an add on to EIAJ 1/2″ reel to reel machines. The Philips N1500 and N1700 cassettes and also Sony U-Matic (both predated Betamax and VHS) used a similar scheme.

[ratinox]

Originally Posted by JSWolf

CD > MiniDisc > CD will result in a loss of data. Any stripping is not 1:1. You can even have a difference going from CD > CD unless you have the correct offset for the drive you are using.

Not exactly. The data transferred to the MD recorder via TOSLINK is 1:1 identical to the data on the CD, assuming the read from the CD player is good. Storage on the MD media itself is lossy if ATRAC is in use (there is a lossless MD encoding format though this came very late in MD's lifetime) so going the other way will incur generational loss. This is the same case as if D-VHS uses lossy encoding which I don't know if it does or not.

Point is, the digital data transfer is 1:1.

SCMS stripping does not affect the data at all. It resets the copy count bit and/or the copyright bit in the status subchannel.

[ratinox]

Originally Posted by Quoth

CD to CD is only 1:1 for CD-ROM, not CD-Audio as audio has no error correction, though mostly the results are better than analogue tape and very many more generations are possible.

Not exactly. CD-DA (Digital Audio) does have error correction but it's only two rounds of Reed-Solomon coding which is sufficient for single bit errors. Full error correction requires three rounds of RS coding. CD-ROM uses three rounds.

You can get bit-perfect copies of CD-DA, down to the offsets and subchannel metadata. The drive firmware needs to know how to do it but it is very possible if the media isn't too badly damaged, and is a sought-after feature for anyone looking to rip their music library. I have owned three such drives: an old Lite-On CD-ROM drive which I think I no longer have, a drive from LG which I picked up specifically for backing up my Saturn and PlayStation libraries (ReDump approved), and a Philips Blu-ray drive as a backup for the LG.

Plus the AccurateRip database exists.